… visual attention model on top of CNN features for unconstrained scene text recognition. This
… In this paper we proposed an LSTM-based visual attention model for scene text recognition. …

… propose a Single Visual model for Scene Text recognition within the … modeling entirely. The
method, termed SVTR, firstly decomposes an image text into small patches named character …

… [13] show visual attention can effectively boost the performance in a LSTM-based … work to
text detection, and build a model for end-to-end scene text recognition with simpler structure. …

… models, which use a DNN with attention model, we investigate a DNN model that can
recognise scene text … The proposed DNN model is named deep attention neural network (DANN), …

… (2017) show that visual attention can effectively boost the performance in a LSTMbased …
text detection and build a model for end-to-end scene text recognition using a simpler structure. …

… a novel transformer model for scene text recognition. Our model is a unified end-to-end text
recog… In this paper, we propose visual-semantic transformer to solve scene text recognition …

… , which is an extended attention mechanism, to the unified text recognition framework to deal
with the task of STR. We propose a five-stage model for scene text recognition, and the AoA …

… In this paper, we present an end-to-end Attention Convolutional Network for scene text
recognition. Firstly, instead of RNN, we adopt the stacked convolutional layers to effectively …

… of text recognition, and propose a novel scene text recognition method to imitate the mechanisms
of saccades and fixations. Second, the model we … in the field of scene text recognition. …

… attention model [24] adopted in [16] only uses the alignment information at the last time step.
Therefore, the state-of-the-art attention-based methods for scene text recognition … attention …